Remote sensing for power grid fuse tripping using ai-based fiber sensing with aerial telecom cables

Abstract

Disclosed are integrated DFOS/DAS systems, methods, and structures that advantageously detecting fuse cutoff blowing events using existing telecom cables. Our systems, methods, and structures employ an exciter to broadcast acoustic signal tracks and evaluated on wooden utility poles within a real-scale testbed, simulating fuse cutoff blowing events in the power grid. A Distributed Acoustic Sensing (DAS) system connected to an optical fiber sensor cable collects a 2D waterfall matrix. A frequency learning model is subsequently used to identify these acoustic events based on the results of frequency analysis.

1 . A distributed acoustic sensing (DAS) method comprising:
 collecting, using a DAS system, backscatter data from an optical sensing fiber;   generating, 2D waterfall matrices from the collected backscatter data;   developing, a frequency learning model to identify different acoustic events; and   determining, using the frequency learning model, different acoustic events acoustically affecting.   
     
     
         2 . The method of  claim 1  wherein at least one of the determined acoustic events is a fuse tripping event occurring on an electrical power facility. 
     
     
         3 . The method of  claim 2  wherein at least one of determined acoustic events is a transformer malfunction event. 
     
     
         4 . The method of  claim 3  wherein the optical sensing fiber is suspended aerially on a plurality of utility poles and at least one of the determined acoustic events is a gunshot. 
     
     
         5 . The method of  claim 2  wherein the DAS system is a coherent detection-based DAS system. 
     
     
         6 . The method of  claim 5  further comprising:
 training, the frequency learning model using simulated fuse tripping events. 
 
     
     
         7 . The method of  claim 6  wherein the simulated fuse tripping events are performed by pole-mounted exciters. 
     
     
         8 . The method of  claim 5  wherein frequency response of collected DAS data differs from its acoustic source. 
     
     
         9 . The method of  claim 6  wherein the at least one determined acoustic event is determined by a 1D frequency response model. 
     
     
         10 . The method of  claim 9  wherein the 1D frequency response model receives as input an extracted phase from complex fiber sensing data, performs a Fast Fourier Transform on a time series of that complex fiber sensing data generating a frequency response and applies the frequency response to a succession of 1D convolutional layers for nonlinear feature learning and fully connected layers to derive probabilities of classification.